1
Neutrality and Political Liberalism
Richard J. Arneson
For Montreal Conference
very rough draft
Is neutrality on the good in some sense an ideal that a just society must embrace?
A flagrantly non-neutral policy such as a state establishment of religion would surely
merit condemnation. For example, we might imagine a society that compels attendance
at Roman Catholic religious ceremonies by law and limits the freedom of Protestants,
Jews, Buddhists, and others to practice their faith as they see fit. Such a policy would be
non-neutral on the good, but would also run afoul of other principles of right that might
have independent appeal for many of us, whatever our views on state neutrality. On the
other hand, we might imagine a society in which human friendship is deteriorating
according to a variety of social science measures. An alarmed government institutes laws
and policies to promote friendship, consisting of a pro-friendship advertising campaign,
attention to friendship in the public school curriculum, and subsidized psychotherapy and
anger management classes for people who see themselves as lacking capacities for
forming and sustaining friendships and are unhappy about that. Such policies would
qualify as non-neutral, if the neutrality norm rules out state action that advances some
controversial conception of good, and if it is controversial that friendship is good.
(Maybe some theorists celebrate the hermit’s life.) Here non-neutrality does not strike
me as bad policy.
Neutrality on the good is a thesis in the theory of moral right, the theory of what
we morally owe to one another.1 More specifically, neutrality is a suggested moral
constraint on justifiable state policies. (Neutrality also forbids some uses of private
coercion, but in this essay I focus on the state policy constraint.) Neutrality is opposed to
perfectionism, regarded as (inter alia) a norm for the guidance of state policy. However,
both neutrality and perfectionism come in different versions, and one might wonder
whether plausible, moderate versions of each view might tend to convergence, that is, to
recommendation of similar practical policies. The first part of this essay argues that the
opposition is robust and the prospects for convergence are dim.
Political liberalism as affirmed by John Rawls might seem to provide strong
arguments in favor of a version of pluralism on the good. The second part of this essay
argues that this appearance is deceptive. Thinking through the political liberalism
approach to social justice issues yields no support for neutrality on the good.
I.
Clarification. First, some clarification of the neutrality ideal is needed. I shall
distinguish three notions of neutrality and identify the neutrality norm with two of them.
Here I am following the practice of neutrality advocates, not claiming to do any original
analytical work. The three notions are:
1. Neutrality of aim requires that no action or policy pursued by the state should
aim to promote one controversial way of life or conception of the good over another.
2. Neutrality of justification requires that any policies pursued by the state should
be justified independently of any appeal to the supposed superiority of one way of life or
conception of the good over another.
3. Neutrality of effect requires that policies pursued by the state should not bring
it about that any controversial way of life or conception of the good is advantaged over
2
others. Nor should state policy bring it about that any adherents of some controversial
conception of the good or way of life are advantaged over adherents of other ways or
conceptions.
The neutrality norm as I shall understand it in this essay combines neutrality of
aim and neutrality of justification. Both requirements must be satisfied.
In passing, I note that neutrality of effect should perhaps not be dismissed too
quickly. One might suppose that if one cares about the values that underlie the neutrality
norm, one should object not only to policies that deliberately aim to advance one
conception of the good over another, but also to policies that bring about as a foreseen
but unintended side effect the advancement of one conception of the good over another.
Suppose that in order to further civil peace, threatened by religious strife, a government
bans proselytizing by Roman Catholics directed at Protestants but permits proselytizing
by Protestants directed at Roman Catholics in a society in which only a small minority
adheres to the Catholic faith. Just assume that the ban just described more effectively
promotes the worthy aim of civil peace than would a different rule that symmetrically
forbids proselytizing by both sides. Social scientists inform us that Protestant sects will
flourish and the Catholic faith will wither under the asymmetrical ban, but to symmetrical
ban would have no such effect. Surely if we care about neutrality on the good at all we
will object to the government policy just described.
Roughly, one might hold that when a government adopts a policy that has the
unintended but foreseen side effect of promoting one way of life or conception of the
good over others, the good that the government aims to achieve by the policy must satisfy
a proportionality requirement. That is, the unintended nonneutral effect must not be
disproportionate, excessive by comparison with the good to be achieved by the policy.
More broadly, public policies must overall be fair in their distributive effects on
individual life prospects, and the fairness standards that are appropriate in this connection
will pay due heed to the values associated with the neutrality ideal.
Having suggested that we might expand our understanding of the neutrality norm
to include some version of neutrality of effect, I ignore the suggestion in what follows. I
follow the custom of understanding neutrality to be the combination of neutrality of
justification and aim.2
The right, independent of the good. A lot here hinges on how one understands
the notion of the morally right that is counterposed to the notion of the nonmorally good,
the neutrality doctrine prescribing neutrality only with respect to controversial
conceptions of the latter. To see this, consider utilitarianism, usually regarded as a
paradigm of a theory of right that is essentially enmeshed in nonneutrality on the good.
Utilitarianism specifies what is good independently of notions of what is morally right,
then specifies the morally right as doing what maximizes the good. However, a
utilitarian perplexed by neutrality might reconceive her doctrine. Given intractable
controversy about what is good and worthy of pursuit in a human life, the neutralist
utilitarian eschews taking a stand on this issue. Instead she brackets this question and
proposes that what is morally right is choosing actions and policies so as to maximize
aggregate preference satisfaction. The maximand in utilitarianism is chosen not with a
view to singling out what is truly good but rather with a view to treating all persons fairly
in the light of the unavailability of an uncontroversial conception of good. Preference
satisfaction or life aim fulfillment is not plausibly regarded as what makes a life good for
3
the person who lives it. Everyone recognizes that preferences and desires can go askew
and be oriented toward what is shallow, trivial, and even base. However, letting the input
into the determination of what should be done be the preferences or life aims of each and
every individual, weighted by their subjective importance to the individual whose aims
they are, is a way of treating all persons with respect and fairness. A utilitarian theory of
social justice might seem especially plausible in this perspective. Assign each person in
society the same number, let each one order her preferences so their total satisfaction
would sum to this number, and affirm that society ought to be organized to maximize
aggregate satisfaction of everyone’s preferences. If it is objected that organizing society
on this basis would not be fair, because not everyone values preference satisfaction the
same and some might regard it as not valuable at all, the reply is that our theory of social
justice is not taking a stand on the good and so is not committed to any view on the value
of preference satisfaction, to which some might reasonably object. Everyone’s
preferences to count for one, no one’s for more than one is a doctrine of right, a proposal
about how to treat people fairly in the light of their deep ethical disagreements, according
to the neutralist utilitarian.
I don’t insist that neutralist utilitarianism is a compelling moral doctrine. Instead
the point is that the doctrine of neutrality on the good invokes a contrast between the
good and the right and as it were places its bets on the theory of the right. The
assumption is that a satisfactory view of right can be worked out without drawing in
controversial commitments regarding the nature of human good, and what neutrality
amounts to depends almost entirely on what is made of the residual practical reason
materials left to concoct a doctrine of moral right once controversial ideas of good are set
firmly to the side.
Viewed in the light of this contrast between the concepts of the good and the
right, the doctrine of neutrality reveals an odd asymmetry. Neutrality insists that state
power not be deployed on behalf of controversial ideas of the good. The appropriate
justifications for the use of state power must be sought elsewhere, and in part, in the
doctrine of neutrality itself. The doctrine of neutrality is doubtless itself a controversial
position, but this is a controversial doctrine within the domain of the morally right, so its
being controversial does not rule out its being acceptable, part of the doctrine of justice
that determines what uses of state power ought and ought not to be made.
The evident odd asymmetry here is that conceptions of the good that are
controversial are ipso facto unsuited to be determiners or partial determiners of what is
just, whereas controversial conceptions of right are not barred from functioning in just
this way. Presumably the background assumption is that a conception of what is morally
right can be controversial, accepted by some intelligent and thoughtful people and
rejected by others, yet still be correct for all that, singled out as uniquely best by the all
things considered balance of moral reasons that bear on the issue. Why this epistemic
asymmetry? Something approaching skepticism about the rational determination of what
is good is assumed, along with robust nonskepticism about the rational determination of
what is right.3 One wonders whether the bar of rational acceptability is being raised and
lowered arbitrarily when one moves from the domain of the right to the domain of the
good, so the standard cannot be met when we are debating what is good but can be met
when we are debating what is right. The suspicion is naturally raised that whatever
skeptical considerations are thought to defeat the project of figuring out what makes a life
4
good for the person who lives it, those same skeptical considerations applied
evenhandedly to the project of figuring out what we morally owe to one another will
bring about a similar defeat. This is just a suspicion, and one that could be defeated by
presentation of arguments that warrant skepticism about the good and fideism about the
right. I confess that in my own mind the suspicion rankles. I have yet to see the
arguments that amount to a plausible defense of asymmetry.
However, it should be noted that neutrality might be defended without appeal to
asymmetry. For example, one might adhere to a theory of moral right whose central
components simply do not hinge on claims about the good. If Lockean libertarianism
ultimately proves to be the correct theory of moral right, then what we owe to one
another does not except perhaps at the margins depend in any way on conceptions of
what goals are most worthy of pursuit and what ways of life are most fulfilling.4 The
Lockean libertarian holds that each person has the right to do whatever she chooses with
whatever she legitimately owns provided she does not thereby harm others in certain
ways that count as violating their natural rights. Each person also has a companion right
not to be harmed by others in any of these certain ways. Moreover, each adult person is
the full rightful owner of herself and can acquire full ownership of material goods
according to principles of first appropriation, voluntary transfer, and rectification.
Finally, the Lockean holds that each of us must always respect—refrain from violating—
all persons’ moral rights as just sketched. It follows that a state would at most be
justified in acting to protect people’s natural moral rights. No state policies that run afoul
of the neutrality norm could be justified, unless people band together and voluntarily
form a state. In that scenario people would be bound to whatever they agreed to, which
might well include nonneutral state policies. The agreement would of course be valid
only among those who freely consent to its terms. The traditional state that imposes its
rule on nonconsenting individuals would be morally permitted to carry out the task only
of protecting people’s Lockean rights, and would not be permitted to impose on people in
order to promote their good, regardless of whether this was done in the name of
controversial or uncontroversial conceptions of good. In this way the Lockean libertarian
position arrives at a position close to neutrality without reliance on any epistemic
asymmetry claims to the effect that knowledge of the good is unavailable whereas
knowledge of the right is available.5 Knowledge of the good, if such there be, would be
simply an idle wheel in the Lockean libertarian political morality.
I don’t regard the line of thought just adumbrated as a defense of the neutrality
norm. I say this even setting aside the question, whether the neutral appearance of
Lockean morality is misleading. Even if a Lockean doctrine of moral right is committed
to neutrality on the good, that observation does not tend to show that neutrality on the
good is justifiable unless Lockean libertarianism is justifiable, which I doubt is the case.
But since neutrality is one small component of many possible doctrines of right, the
decisive evaluation of neutrality must wait upon the completion of the decisive
assessment of all remotely plausible doctrines of right in which neutrality might figure.
Even if neutrality looks dubious regarded on its own, it might be a necessary element in a
political morality that looks impressive overall. If neutrality is a beast, it is a beast that
will be difficult to kill by argument.
Neutrality regarding controversial conceptions of good. The neutrality norm
requires the state to be neutral only between controversial claims about what is good.
5
Which are these? A conception of the good might be controversial in a purely descriptive
sense. In this sense, it something is actually disputed, it is controversial. I suppose a
friendly interpretation of the neutrality norm takes a claim about the content of human
good to be controversial if it is normatively controversial. There are reasons for and
against the claim, so it makes sense that the claim is controversial.6 The claim has
features that make it apt for controversy. I suppose also that a claim about good can be
controversial in this normative sense even when it is the case that it is ultimately
ascertainable whether or not the claim is true or best supported by moral reasons, all
things considered. There is a truth of the matter that in some sense, maybe with
difficulty, maybe only in the long run, is amenable to discovery. That can be so, yet the
claim is controversial, and given our epistemic situation, should be controversial, to a
degree, for now.
There is a puzzle here, however, to which I keep recurring in this essay. If there
are some reasons in favor of a view, say Buddhism, and some reasons in favor of an
opposed view, say Hinduism, and no compelling reasons on either side, how does the
choice between these doctrines get to be legitimately controversial? If the situation is as
described, shouldn’t we all agree there is something to be said for each side and no
decisive reason to favor either one? Perhaps we should say that an issue is descriptively
controversial if actually disputed and normatively controversial if there is some rational
basis for controversy, even if not a decisive basis. Being normatively controversial then
admits of degree. At the limit, all reasoners responding fully to the reasons there are will
agree (even if just to say there is no superior merit in any contending position). Just shy
of that limit, there may still be controversy among pretty good reasoners, and as one
raises the standard of normative controversiality, one gets closer to the limit.
The neutrality norm advocate would render her position implausible if she held
that anything actually disputed is genuinely controversial and would be gutting her
position if she insisted on raising the standard for normative controversiality to the limit
where nothing would actually be disputed. There is as it were a Goldilocks issue here; a
plausible neutrality norm will insist neither on too much reasonableness nor too little in
setting the bar for being normatively controversial.
What neutrality comes to in practice will depend on what counts as a
controversial conception of good or claim about the good. I assume that advocates of the
neutrality norm would agree that is a controversial claim about the good that the goods of
sexual pleasure, sexual intimacy, and sexual friendship that we suppose are available in
heterosexual relationships and heterosexual sexual activity are equally available in
nonheterosexual relationships and sexual activity. These important human goods that a
man and a woman can gain from sexual encounters can equally be gained from same-sex
sexual encounters.
This claim is clearly controversial in the descriptive sense. Many Christians and
others as a matter of fact reject it. More to the point, the claim is normatively
controversial; it is a claim that is controversial among ordinary, reasonable people who
are not substandard in cognitive abilities. People offer substantial arguments in its
support. People who find the idea of gay sex degrading or evil sometimes appeal to
Divine Command to justify this response, so the issue is then entangled in further
controversial issues concerning the interpretation of the meaning of specific Scriptural
passages, the interpretation of Scripture as Divine Revelation, the view of Divine
6
Commands as constituting what is good as well as what is morally obligatory, the
coherence of theism, the arguments for God’s existence, and much else. Intelligent
people can be found on opposed sides in such disputes.
Despite its controversiality, I find it incredible that the claim that same-sex sex is
good is false. It strikes me as evidently true and a suitable basis for public policy in a
diverse democracy. I don’t propose to argue for the goodness of same-sex sex here; I just
invite the reader to consider the issue and ponder whether neutrality regarding
controversial conceptions of good in its application to this question looks like a stance
that justice allows. I think not.
One might resist this plea for nonneutrality on the ground that what is required is
that the state protect the rights of all persons to conduct their lives as they see fit
according to their own values and norms so long as they don’t harm others (in certain
ways that violate their rights). Taking a stand for or against the goodness of sexual
activity of any sort is neither here nor there, one might hold. So the claimed requirement
that the just state should take a stand on this matter does not put any pressure on the
advocate of the liberal neutrality. What the just state does is keep its mouth shut on this
issue, and that’s consistent with a strict neutrality policy.
We should not buy this line of thought. Merely allowing people the freedom to
pursue their good as they see fit may not give them a fair opportunity actually to lead
good lives, because they lack the resources necessary to achieve the reasonable goods
they seek. Nor is this problem remedied by adding to one’s theory of the right a
provision that ensures that people will have a fair opportunity to gain resources or even a
guarantee that they will have access to some threshold amount of multi-purpose resources
suitable for gaining diverse goods. The problem is that the resource share that is fair for
an individual to have cannot be settled independently of considering the individual’s
particular traits and how those traits would interact with given resources in the
individual’s actual circumstances to generate opportunities to achieve a good quality of
life. What counts as a good quality of life depends on what is valuable in human life, not
merely on what is currently fashionable or revered. On this view, what we owe to one
another depends inter alia on what is objectively valuable and what it would take for each
of us to get a fair amount of it.
Reverting to the example of same-sex sex, I claim that merely ensuring that each
man and woman has the freedom to pursue romantic partners who are either men or
women does not guarantee adequate opportunity for romantic fulfillment to all regardless
of their sexual orientation. Perhaps instruction to all school children is needed to bring
about a social atmosphere that promotes real as opposed to merely formal freedom in this
sphere. A society that passes laws forbidding racial discrimination does not merely aim
to prevent racially discriminatory behavior. The larger aim is to change the hearts and
minds of men and women, so they come to be no longer prone to racial animus and
prejudice. To achieve the larger aim might require additional policies whereby the state
declares itself for a racially harmonious society. In much the same way, bringing it about
that--so far as is feasible consistent with other justice goals--all individuals have adequate
opportunities for sexual and romantic fulfillment may require not merely laws protecting
people’s freedom and privacy but also nonneutral state policies whereby the state
declares itself to be for the (controversial) good in this area.
7
Perfectionism, extreme and moderate. A neutrality norm with its scope limited
so it applies only to controversial conceptions of the good and a perfectionist ethic that
recognizes pluralism and limited commensurability might seem to be heading toward
convergence. We might wonder if there is room left for significant, consequential
practical disagreement when the two principles are qualified and refined and accepted
finally only in their most plausible versions.
A perfectionist ethic holds that one morally ought to bring it about that human
good is promoted, at least up to a point (and within the limits of whatever constraints and
options on such promotion should be accepted). For any individual person, human good
is greater when one’s life goes better for one rather than worse, and a life goes better for
one to the degree it includes more objectively valuable things, the entries on a so-called
Objective List.7 As a life gains more and more of such goods, each weighted by its
intrinsic importance, the life resembles more closely the perfect human life. The good in
question here is prudential good, the good an individual seeks when she is seeking her
own interest or advantage.
A simple perfectionist ethic would have it that all moral requirements are
subsumed in the requirement to promote the objectively good; this yields a version of act
consequentialism. The aim of promotion might also take a variety of forms: one morally
ought to maximize the aggregate sum of human good, or the average level, or one ought
to maximin, or prioritize, or equalize, or one ought to bring as many as possible to the
sufficient or good enough level, or one ought to give priority to gains for those below the
sufficient level, and so on. A more modest perfectionist ethic would also accept some
deontological constraints and options alongside its requirement to promote the good.
However, even on the modest version, a perfectionist ethic holds that to some
considerable extent, what is morally right depends on claims about what is objectively
good for people and about how such good can best be promoted.
Perfectionism might assume an extreme or a more moderate form. Extreme
perfectionism affirms full commensurability. For any different ways that people’s lives
might go, if the facts are fully specified, then there is always a determinate answer to the
question, which possible life of those the person might have led would have been best for
her, and by how much, and how much better or worse would a given possible life of this
person have been, compared to specific lives that other people might have led. In
principle there is a definite answer to the question, for any combination of types and
amounts of good in a person’s life, what is the total of well-being that life provided her.
On an extreme view, goods of different types can be ranked against each other, and
numbers can be assigned to persons lives, registering how good it was for the person to
lead that life, that are cardinally interpersonally comparable. Full commensurability
might be thought to arise by virtue of the fact that the apparent variety of human goods
reduces to a single good, which admits of definite measurement. Or full
commensurability might be possible even if there are several goods, which can be
assigned determinate weights, so for any combination of different goods, in principle the
sum of good can be calculated,.
Moderate views affirm partial or limited commensurability. There is a range of
such views. Some might countenance only intrapersonal comparisons of well-being.
8
Some might countenance ordinal not cardinal comparisons, whether interpersonal or
intrapersonal. I want to highlight moderate views that countenance limited, gappy
cardinal interpersonal commensurability. If two persons’ lives are otherwise identical,
but one has some of good X and the other has some of good Y instead, then holding fixed
the quantity of X the one has, there will generally be some amount of Y, such that if the
other person has less than that amount of Y, her life has less well-being than the other,
and there will be some amount of Y, such that if the person has more than that amount,
her life has more well-being than the other. Between the higher and lower amounts, the
one is neither better off nor worse of nor exactly as well of as the other. This is still just a
rough illustration, because one must also take account of interaction effects (if one has
buttered popcorn, that might be better for one than the sum of the amount of good one
would get from having just the butter and the amount one would get by having just the
popcorn. Perhaps one needs to insist on fine-grained specifications of goods that
incorporate all interaction effects.)
Perfectionism versus neutrality. Perfectionism then immediately conflicts with
neutrality, regarded as a constraint on acceptable state policy. Perfectionism says that
what is morally right is to promote the good, and bring about its fair distribution, within
whatever moral deontological constraints ought to be accepted. This norm is to regulate
choice of public policies and laws enforced by government as well as choices by
individuals about what courses of action to adopt. The correct conception of good
largely fixes the substantive content of morally right public policies and laws according
to perfectionism. However, neutrality opposes perfectionism on just this point.
Neutrality holds that it is morally wrong to use state power with the aim of advancing
some controversial conception of human good and morally wrong to impose on people in
the form of laws and public policies that could be justified, if at all, only by appeal to
controversial conceptions of good. If the correct conception of human good is
controversial, perfectionism affirms exactly what neutrality denies. The component of
the neutrality norm that I am saying conflicts with perfectionism is neutrality of
justification.
One might resist this conclusion from various standpoints. One might suppose
that the more moderate one’s perfectionism, and the higher the standard that must be met
if a conception or claim about the good is to count as controversial according to the
neutrality norm, the less room is left for serious disagreement as to what state policies in
this area are morally acceptable. Broadly speaking, this claim is true. But two caveats
should be added. First, the neutrality advocate does not have unlimited freedom to raise
the bar that must be met for a dispute about what is good to qualify as genuinely
controversial. Beyond some point, this strategy simply amounts to abandoning the
neutrality norm as usually conceived. Second, as the example of state endorsement of the
value of same-sex sexual activity illustrates, a claim about what is one alternative form of
the good life can be just as controversial as claims about what is uniquely good.
Perfectionism as characterized just above says state policy should promote the
good within acceptable deontological constraints, and maybe conformity to neutrality is
one such constraint. Also, perfectionism as just characterized says state policy should
promote the good and bring about its fair distribution, and perhaps fairly distributing the
good requires adherence to the neutrality norm. Maybe these two possibilities come
down to pretty much the same.
9
The thought that neutrality is a fundamental deontological constraint on
promoting human good via state action suggests that there is something special about
state action that triggers the constraint in this special context. One possibility here is that
what makes state action special is that it is coercively enforced; behind announced
governmental edicts there are sanctions—the police, the law courts, jail, fines, prison,
public condemnation, the gallows, and so on. However, coercion per se cannot be the
pivotal concern, because state policies could be nonneutral in ways the neutrality norm
deems objectionable even if they are cajoling and not coercive. Another possibility is that
in the nature of the case, the state at least in modern times represents itself and moreover
ought to represent itself as speaking in the name of all the members of society and doing
so in ways that all members have reasons to accept. This consideration I take to
introduce the core concern of political liberalism, examined in part II of this essay.
A perfectionist view that recognizes a plurality of distinct and independent goods
and countenances only limited commensurability across the goods it recognizes will still
oppose the neutrality norm for a simple reason not yet mentioned: neutrality includes the
requirement of neutrality of aim.
A perfectionist doctrine that is pluralistic and envisages only limited
commensurability across values might well end up recommending policies that violate
neutrality by promoting one controversial value over another even if the favored value is
not deemed superior to the disfavored value. How this would come about depends on the
particular principles of right the perfectionism qualifies. For simplicity assume a simple
maximizing ethic. We are aiming to maximize aggregate human good, summed across
persons, or summed in a prioritarian way that gives extra weight to achieving gains for a
person, the worse off the person would otherwise be over the course of her life.
Perfectionism of this sort might recommend as morally right a state policy that subsidizes
opera and not jazz even though there is no ground for regarding the one musical art form
as superior to the other. Contingencies might nonetheless dictate promoting opera alone.
There might be economies of scale in the enterprise of encouraging the production and
consumption of musical performance, so more good for people is gained by channeling
an extra ten million dollars into opera to top off an initial ten million dollar grant rather
than balancing the opera subsidy with an equal subsidy to jazz. There might simply be
many more people who will benefit from the promotion of opera than would benefit from
the promotion of jazz, given the distribution of tastes, so again, more value is gained by
concentrating resources on opera at the expense of jazz. If we are prioritarians, we will
hold that the correct choice of policy depends on how well off or badly off the potential
beneficiaries would otherwise be, so if he opera lovers are poor peasants and the jazz
lovers are rich bankers, there is an extra reason to promote what aids the former. The
envisaged nonneutral policy would violate neutrality of aim but not neutrality of
justification.
The example just illustrates a general point: merely affirming that many different
types of thing are objectively valuable and such that getting or achieving things of the
type enhance people’s well-being does not necessarily tend to bring perfectionism and
neutrality into convergence. Nor does moderate perfectionism (committed only to limited
commensurability) necessarily imply only policies that are compatible with neutrality or
close to that. A pluralistic moderate perfectionism is still the enemy of neutrality.
Whether these possible disagreements materialize in practice depends on the
10
circumstances we face, but my own hunch is that in actual and likely circumstances the
perfectionist advocate and the neutrality advocate will disagree sharply and across the
board.
II.
Political liberalism and neutrality. Many theorists nowadays who endorse
some version of the neutrality doctrine find its most philosophically sophisticated
development and elaboration in the ideas of political liberalism first articulated by John
Rawls.8 Accordingly I turn to political liberalism for the assistance it can offer for the
assessment of neutrality.
Rawlsian political liberalism starts with a problem: Can there be reasonable
agreement among the members of a diverse liberal society on a philosophical account of
justice that indicates why the basic institutions of the society are just, to the extent they
are that, and where these institutions fall short, what changes would render them
acceptable? On the face of it, the answer seems to be No. Given what Rawls calls the
fact of reasonable pluralism, unless there is tyrannical restriction of freedom of thought to
secure unanimity of opinion that could not conceivably be acceptable, the members of a
liberal society will fan out into allegiance to opposed comprehensive moral doctrines. A
comprehensive doctrine provides answers to a wide range of ethical questions about how
we ought to live, what is worthwhile and valuable in human life, what rights people have
by way of fair treatment from others, along with corresponding duties, what character
traits should be instilled and emulated, what shape the basic institutions of society should
take, and so on. A comprehensive doctrine might also include views about the place of
humans in the universe and views on philosophical and metaphysical issues concerning
the natures of things. Not everyone has opinions that amount to a comprehensive view,
but many do. Traditional religions tend to provide large worldviews amounting to
comprehensive views. Reasonable people will differ widely in their allegiances to
comprehensive doctrines, and the disagreements will be stable and enduring. So far as
we can see, we are never going to agree about these matters.
Given the fact of reasonable pluralism, it seems that any philosophical conception
of justice will be contested, and rejected by some persons on the basis of their reasonable
comprehensive beliefs. Unanimous agreement on principles of justice will forever elude
us. But then we seem stuck in a swamp of incoherence. It then seems we together could
never have reasonable confidence that our basic institutions and practices are just, or
making good progress evolving toward justice. Moreover, it might seem impossible to
fulfill what Rawls takes to be a fundamental moral constraint on acceptable employment
of state power, the legitimacy ideal: One morally ought not to use state power in ways
that coerce individuals or affect their interests except when these state policies are
justifiable by reference to principles that no one could reasonably reject.9 But given the
fact of reasonable pluralism, any principles some might reasonably affirms, others will
reasonably deny.
Rawls’s solution to this puzzle is by now well known.10 A reasonable person,
aware of the situation just described, will not attempt to secure unanimous agreement
from her fellow citizens by making sectarian appeals to principles some will reasonably
reject. Instead reasonable people will strive to satisfy the legitimacy ideal by working out
whether public reasons, shareable by all, will single out principles of justice for the
11
regulation of the basic structure of society and of the conduct of individuals that are
reasonably rejectable by none. If this enterprise is successful, we will have identified a
free-standing module consisting of fundamental moral principles and their intuitive
rationale. These principles with their intuitive rationales will be acceptable to all
reasonable persons. They represent (what should be) our rock-bottom core commitments,
the shared basis for social unity, the standard of social justice.
Isn’t this sort of unity ruled out by the initial formulation of the problem? Since
reasonable persons give their allegiance to opposed comprehensive principles, any
principles of justice affirmed by some will be rejected by others. One way to thread the
needle here would occur if some moral principles turn out to be common to all
comprehensive doctrines even if many principles are not. Rawls sometimes does speak
in this connection of an overlapping consensus, which is the idea that a political
conception of justice might consist in a set of conclusions derivable from different and
opposed comprehensive doctrines. No two comprehensive doctrines agree in the
premises of the derivation, but by different and conflicting paths all get to the same
destination, the same conclusions.
Overlapping consensus is one possibility, but it’s not needed for the requisite
agreement. Alternatively, given that we are aware of our comprehensive doctrine
disagreements and seek shared rational principles despite this, we seek public reasons
that all who are reasonable will affirm in the light of this conception of the task—to see
what we can agree on, given we disagree intractably about so much, “we” being rational
persons who seek to cooperate with others according to norms no persons with just these
aims could reject. We seek agreement just for this limited purpose and eschew candidate
principles and rationales not needed for this purpose. The public reasons and linked
principles we affirm at the end of this exercise might not include any principles from
anyone’s comprehensive doctrines, just freestanding moral ideas tailor made for inclusion
in a political conception of justice—a conception that can attract rational unanimous
agreement despite deep pluralism and disagreement in belief.
Nothing guarantees that such a search will be successful, but nothing guarantees it
won’t be successful.
The key is supposed to be that the reasonable person in a diverse democracy can
negotiate two tiers of belief, comprehensive and political, and two kinds of reasons,
private and public. More on public and private reasons in a moment.
The relationship between political liberalism as just described and the neutrality
doctrine is transparent. At least, political liberalism might appear to imply neutrality of
justification (I shall deny that this appearance is correct). The principles we should
affirm as a political conception of justice are principles supported by the balance of
public reasons. But any controversial claim about what is good and worthwhile in human
life will be unacceptable from the standpoint of some reasonable comprehensive doctrine,
so no such claim can be advanced as a public reason shareable among all persons and no
such claim can have any influence in shaping a political conception of justice.
Notice that political liberalism so construed vindicates the neutrality doctrine
without any invocation of the odd epistemic asymmetry between conceptions of the food
and conceptions of the right. No such asymmetry plays a role in political liberalism. The
operative contrast is instead between a comprehensive ethical doctrine, a worldview that
guides all or at least many components of practical reasoning, and a political conception
12
of justice, a stripped down minimalist moral doctrine that answers to the narrow question,
what principles can be affirmed by all rational and reasonable11 persons as suitable for the
regulation of society (basic institutions and the conduct we strictly owe to one another) in
the light of the evident fact that we shall never rationally and reasonably agree on
comprehensive doctrines.
Empirical versus moral disagreement. As stated, the legitimacy ideal seems to
imply that being coerced or imposed upon in the name of moral principles you do not
accept is morally obnoxious in a way that being similarly coerced or imposed on in the
name of empirical claims you do not accept would not be. I wonder about that.
Imagine that I am the owner of a factory that is subject to strict government
regulation constraining its operation. The point of the regulation is to prevent me from
contributing excessively to global warming or some other long-term environmental
threat. I chafe at the regulations, which I claim to be in sharp conflict with the ideal of a
society of free and equal persons. Now imagine two variants of the story. In the first, I
reasonably reject the moral principle that is invoked to justify the regulation. I deny that
present property owners are required to be stewards of the earth, are bound by obligations
to future generations, or anything of the sort. In the second variant, I accept the moral
principle invoked to defend the regulations in question, but I reject the empirical claims
that, combined with the principle, yield the conclusion that I am violating moral duties to
future persons by the manner of operation of my factory and ought to be forced by
suitable regulation to comply with those duties. It is simply not clear to me that the two
variants differ with respect to the prima facie wrongfulness of coercing me or imposing
on me.
In both cases the regulations imposed on me are unjustified just in case the claims
of moral principle and the claims of fact offered as justification are incorrect and there is
no substitute justification that would be correct. (Let’s set aside the version of the latter
case in which, unbeknownst to anybody involved, a correct justification for the
regulations exits, but it is sheer luck that this is so, since nobody is in a position to appeal
to it.) If either my moral principle objection or my empirical fact objection to the
imposed regulations is correct, I am being wronged by their imposition.
However, suppose my objection in either case is not correct, but is reasonable
according to some epistemic standard of reasonableness. The question then arises, how
good is the standard? I’m inclined to say that if I have any right not to be imposed on in
ways that strike me as objectionable, what offends is not just being imposed on in the
name of some controversial view, moral or empirical, but rather being imposed on in the
name of some view that is incorrect (and that cannot be justified by appeal to some other
correct view). Alternatively, one might understand this right in a relativized way: each of
us has a right not to be imposed on in ways that cannot be justified by appeal to where the
balance of reasons available to us points (as indicating what is most likely to be correct
on the basis of presently available evidence and argument).
Problems with political liberalism. The ideal of public reason conjures up the
picture of a harmonious democratic society marked by toleration, diversity, and
commitment. All citizens are united in their commitment to principles of right that
justify the political constitution and basic institutional arrangements. The reasons that do
this justifying work are public reasons shared by all. Beyond the basic arrangements,
citizens fan out into allegiance to diverse and conflicting philosophical and religious
13
commitments and opposed beliefs about what constitutes a good and worthy life. These
various beliefs are justified by private reasons, which all sense would be unsuitable if
proposed as justifications for public policies binding on all. Each believes herself to be,
and perhaps is, perfectly rational in both he public and private commitments, and
recognizes that her fellow citizens believe themselves to be, and perhaps are, perfectly
rational in the same way. I fully believe and endorse my conception of the good, my
nonpublic moral beliefs, and my metaphysical and religious worldview. To endorse these
beliefs is to regard them as correct responses to the reasons there are. Moreover, I see
myself as one rational citizen among others, who are committed to conflicting and
opposed beliefs. The line between public and private reasons is the line between reasons
all can share and reasons that some will accept and others will not.
However attractive this picture, its description of many rational persons holding
equally rational but opposed beliefs is incoherent. If the members of society lived in
isolated valleys, each perhaps could be adequately responsive to all of the evidence she
has, yet adhere to beliefs that contradict what others believe. But that is not the picture we
are given. The gap between public and private reason is one we are supposed to be able
to acknowledge together. But how can I continue to believe in Roman Catholic theology
on the basis of certain reasons while acknowledging that you are perfectly rational to
believe in an opposed theology that conflicts with Catholicism on the basis of other
reasons? One or both of us must be failing to respond fully to the reasons there are.
Either your reasons outweigh mine, or mine outweigh yours, or neither set clearly
outweighs the other. In the third case, I should withdraw my belief, and say that so far as
we know, either my theology or yours (or neither one) might be correct. What I cannot
sensibly do is think that we could both be rational to adhere to different and opposed
beliefs on the basis of what is now the same body of evidence, once we have shared our
reasons. Of course, your standards for “weighing” and assessing reasons might be
different from mine, but then the same problem arises again, with the same outcome. We
cannot rationally believe that the members of our diverse democratic society perfectly
rationally affirm conflicting beliefs, except in so far as we think that some have evidence
of which others are unaware. Moreover, in so far as we have reason to think that others
hold evidence or have articulated reasons we lack but that bear on the correctness of the
beliefs we hold, we should seek out and assess that possibly decisive counterweight to
our current beliefs.
This criticism of the political liberalism ideal might not be devastating, or even
troubling. Perhaps the advocate of political liberalism can concede that at the limit of
inquiry, when all evidence and arguments are on the table, all persons not making any
cognitive errors would agree, both in science and ethics. However, short of that day,
evidence and arguments are unevenly distributed across persons. Since the evidence I
have is not the same as the evidence you have, it is possible that we could disagree, even
though neither of us is in error. Since the ideal limit of inquiry either will never be
reached or is extremely far off in the future, the ideal of fully rational and reasonable
persons continuing to uphold conflicting comprehensive doctrines is not threatened with
incoherence.
The same problem pops up in another place, however. What it is rational for me
to believe now regarding what is right and good depends on the arguments available now,
not arguments and theories and principles that will not be conceived and articulated until
14
some future time. So consider a provisional present-oriented version of the ideal limit of
inquiry. This is roughly the idea of the best that one can do now with respect to the goal
of arriving at ethical beliefs best supported by reasons. Being fully rational, and seeking
to do the best we can to arrive at reasonable ethical beliefs (in the wide sense of ethical
the is involved in the notion of a comprehensive doctrine), we will tend to converge, even
if the views we converge on are largely indeterminate and leave many questions
unresolved. Once again, the political liberalism ideal of fully rational persons holding
starkly opposed ethical beliefs, aware of each other’s positions, and affirming each
other’s rationality, provides a stirring but incoherent picture of mutual toleration.
Of course, putting the matter in this way just emphasizes that in fact the political
liberalism ideal is not that of fully rational persons holding opposed beliefs but rather of
reasonable persons doing that. A reasonable person here is above all one who is disposed
to treat people in whatever ways morality requires, to cooperate with others on fair terms.
This is the idea of reasonableness that is invoked when one says to someone who is not
bargaining in good faith, but rather using hardball tactics to gain as much as he can
without any regard for the interests of others, “Be reasonable!” A reasonable person in
this sense is assumed not to be completely incompetent at practical reason, but need not
be fully or perfectly competent. A person’s adherence to a comprehensive doctrine is
reasonable provided she flexibly adjusts the doctrine in response to new experience and
evidence and strives to render the views she accepts consistent and plausible. Such a
person manifests practical reasoning ability at least at some threshold “good enough”
level. However, this notion of reasonableness allows that one can be fully reasonable
while still making cognitive mistakes and mistakes in reasoning in arriving at one’s
particular, detailed comprehensive doctrine.
In this connection we might contrast Rawls’s notion of a public reason, shareable
by others, and fit to be an input into a decision process that leads to moral principles
constituting a political conception of justice that no one can rationally reject, with the
implicit counterpart idea of a private reason. In deciding on one’s comprehensive
doctrine, private reasons suffice. A private reason is a reason for me that I recognize is
not necessarily a reason for everyone else or for that matter anyone else. For example,
one might take as rock-bottom in one’s epistemology the basic claim that a God along the
lines of traditional Christianity exists. This claim is foundational in one’s set of beliefs.
One might take the claim that God exists to be a basic belief, a reason then that constrains
and shapes the rest of one’s belief set, while acknowledging that someone else might take
as rock-bottom a quite different Hindu or other conception of God, or deny that any such
conception of a deity is to have any sort of privileged position in one’s epistemology. If
one takes a relaxed stance on what it is to be reasonable, so that being reasonable
involves being responsive to reasons but not necessarily fully responsive as an ideal
rational agent would be, it is not at all surprising that one then affirms as Rawls does the
possibility, and even the evident fact, that reasonable people affirm opposed
comprehensive doctrines and will continue to do so.
If one wants to characterize an overlapping consensus on a political conception
of justice that includes among the overlapping views comprehensive doctrines formed by
religious faiths, the notion of the “reasonable” needs to modest in its requirements of
cognitive competence and rationality.
15
If one goes this route, what then seems deeply objectionable to me is any
suggestion that what persons seeking a political conception of justice formed by moral
principles that no one could reasonably reject for the purpose could reasonably assent to
is constrained by the comprehensive doctrines that fit the characterization just given. If
there is a constraint here, it has to go just the other way. Whatever emerges from the
procedure of rationally assessing and weighing public reasons until one fixes (if one is
successful) on principles that no one could rationally reject should serve as a constraint
on acceptable comprehensive doctrines. In weighing public reasons to arrive at
nonrejectable principles, it would be a mistake to hold the process hostage to the
comprehensive beliefs people hold that need satisfy only meet a relaxed standard of
semirationality. If the legitimacy ideal is interpreted as holding that state power should
only be used in ways that are justifiable by reference to principles that no one could reject
on the basis of semirational comprehensive doctrines, the legitimacy ideal is being
stretched too far. If one went this route, one would then open up a gap between
legitimacy and justice. Conforming to the legitimacy ideal might require jettisoning
some principles of justice that are best supported by moral reasons, shareable public
reasons, and therefore not rationally rejectable, because the principles would conflict with
some aspect of some comprehensive doctrines that “reasonable persons” (reasonable by
Rawls’s relaxed standard) might affirm, even though these would not be affirmed by fully
rational persons. Pushed this far, the legitimacy ideal is illegitimate.
If on the other hand we tighten up the standards for rationality, meaning that we
count as a rational person in the domain of practical reason, one who responds fully to the
reasons there are and accepts all and only principles that fully express or incorporate
those reasons, the legitimacy ideal is unexceptionable. It asserts: We should deploy state
power only in ways that are justified by moral principles that none could rationally reject.
These are the principles best supported by the moral reasons there are. (If some principles
are tied for best, then reason is indifferent as to which one we act on. If coordination
problems arise here, the coordination issue will tip the balance toward whichever of the
tied for best principles is currently accepted or closer to being accepted and governing
people’s conduct.)
By the same token, one should have no quarrel with Rawlsian political liberalism
once one emphasizes that public reasons trump private reasons (if the idea of a reason
private to myself and not also a reason for anyone is coherent) and that the process of
discovery of the candidate political conception of justice best supported by moral reasons
is not constrained to rule out of court candidate moral claims merely on the ground that
they are rejectable from the standpoint of some semirational comprehensive doctrine that
is not guaranteed to be fully rational. Moreover, once one gets this far, one sees
immediately that the discovery process for the morally favored political conception of
justice might well draw in controversial conceptions of the good provided they pass
muster at the bar of public reason. It seems to me overwhelmingly likely, and more
importantly not ruled out by any arguments we should accept, that conceptions of human
good that are anathema to some comprehensive moral doctrines that are reasonable
enough to qualify as included in the consensus that a political conception of justice
acceptable in the Rawlsian framework must attract, are nevertheless fully defensible at
the bar of public reason and fit to shape the content of the conception do justice we
16
should embrace. Some controversial claims about the good are correct. They partly
determine what is just.
This line of thought can be rephrased in terms of what is owed to people by way
of toleration of their comprehensive doctrines, given that they vary by degree in their
practical reason capabilities and in their skill and diligence at exercising them. Being a
partly rational agent, I would say following Kant, I best express my nature as a rational
agent by conforming to principles best supported by the moral reasons there are. Being a
rational agent, and possessing that moral dignity, I have a right not to be treated by
others, affected by their actions, or ignored by them, in ways that are not justified by the
moral principles best supported by moral reasons. But this moral right cannot be upheld
when it is given a subjective construal. I have no moral right that people treat me only in
ways that are justifiable according to moral principles that appear to me to be rationally
compelling. For I might fail to discern some moral reasons that are valid and imagine I
discern some moral reasons that are not there to be seen.
A threshold of reasonableness? One possibility here is that being “reasonable”
in this context is a threshold notion. To be reasonable is to be reasonable enough, to have
met a sufficient standard. This good enough standard might be variable in its stringency,
depending on what is at stake for the choice at issue. But even when the stakes are very
high, as when the justification of public policies and state-enforced laws is being
appraised, to be reasonable it is not required that one be fully rational, make no mistakes,
not even subtle errors, in reasoning and in weighing evidence, be fully and adequately
responsive to the reasons there are that bear on the choices on is making.
The idea then of political liberalism would be that if someone’s beliefs meet a
threshold standard of reasonable belief foundation, then one ought not to use state power
to impose on him when those uses of state power cannot be justified to him from his
standpoint. The principles that legitimate the use of state power in ways that affect a
person are principles that she cannot reasonably reject. I suppose reasonable rejection
might arise in one of two ways: either the person reasonably (that is: at or above the
threshold level of reasonableness) affirms principles and moral views and conceptions of
the good such that there is no sound justification from the person’s own set of beliefs to
the public principles in question, or there is, but the person reasonably fails to spot it.
Jeremy Waldron gives appealing expression to the ideal of political liberalism
characterized this way. He asserts that “the liberal individual confronts his social order
now, demanding respect for the existing capacities of his autonomy, his reason, and his
agency.” William Nelson interprets political liberalism as committed to the idea that
“principles are adequate only if they should be accepted by all reasonable persons on the
basis of beliefs and values they already hold” (rather than on the basis of beliefs and
values they would hold after ideal reflection with full information). On Waldron’s view
especially, a morally proper respect for persons requires respect for their actual exercise
of their rational agency capacities. These might be limited, but then, so are everybody’s.
none of us is a rational archangel. Hence, respect for persons requires respect for their
rational agency and this respect must manifest in some deference to their considered,
deeply held values, however misguided they might seem to me or to you.
However, once we accept that being less than fully responsive to the reasons there
are qualifies as reasonable enough, there is no limit to the substantive badness of the
beliefs and values that reasonable persons might already hold. So there is no limit to the
17
substantive badness of that to which this version of political liberalism bids me to defer.
Deference to excusable, understandable, there-but-for-the-grace-of-God-go-I error is
morally mistaken when anything morally consequential is at stake. As a being that
possesses some rational agency capacity, I am at the deepest level already committed to
allegiance to the right and good, as these notions are singled out by the balance of moral
reasons. Deferring to my mistakes, errors, and prejudices when these are leading me
seriously astray is not respecting me as the limited partially rational agent I am. So far as
I can see, this point holds for controversial ideas of the good as well as for controversial
ideas of the right.
NOTES
1
. My understanding of the neutrality norm owes a lot to works by Charles Larmore and Joseph
Raz. See Charles Larmore, Patterns of Moral Complexity (Cambridge: Cambridge University
Press, 1987); also Larmore, The Morals of Modernity (Cambridge: Cambridge University Press,
1996). See also Joseph Raz, The Morality of Freedom (Oxford: Oxford University Press, 1986).
Larmore acknowledges a debt to two earlier works, Ronald Dworkin’s essay “Liberalism,”
reprinted in his collection A Matter of Principle (Cambridge, MA: Harvard University Press,
1985) and Bruce Ackerman, Social Justice in the Liberal State (New Haven: Yale University
Press, 1980).
2
. Adding an extra controversial element to the neutrality norm would make arguing against it
easier. Since my aim is to raise doubts about the acceptability of neutrality, considering a weaker
version of it makes sense.
3
In his review of A Theory of Justice, Thomas Nagel observes that he interprets Rawls “as saying
that the principles the principles of justice are objective and interpersonally recognizable in a way
that conceptions of the good are not.” Nagel, “Rawls on Justice, Philosophical Review 82 (1973),
228.
4
On Lockean libertarianism, see Robert Nozick, Anarchy, State, and Utopia (New Yorks: Basic
Books, 1974). See also A. John Simmons, The Lockean Theory of Rights.
5
. I say the Lockean libertarian position is “close to neutrality” because this position does not
seem actually to require adherence to the neutrality norm. For example, suppose that it turns out
that the state can best bring about general compliance with Lockean rights by promoting a brand
of religious faith. Coercing people to embrace one faith rather than another or none would itself
be a violation of the Lockean rights of the coerced. But a state establishment of religion need not
be coercive. Suppose the Lockean state engages in advertising and other forms of proselytizing
designed to encourage adherence to the tenets of evangelical Christianity in order to encourage
compliance with Lockean rights. This violation of neutrality of aim looks to be unexceptionable
from the standpoint of Lockean morality. One might worry that the state would be forcing people
to pay taxes to support a state establishment of religion of which they disapprove. How could this
forcing be morally permissible? First, I am not sure this difficulty is greater for coercive taxation
to support a religion than it would be for coercive taxation to support the maintenance of a police
force or a similar device for securing compliance with Lockean rights that does not trench
against the neutrality norm. Second, one could imagine cases in which the state gains revenue
without reliance on coercive taxation and uses the permissibly gained revenue to proselytize for a
particular religion with a view to carrying out the unproblematic state function of protecting
people’s rights.
18
6
. My descriptions of moral claims and claims about the good life assume a cognitivist
framework in which such claims can be true or false. I suppose some such framework is common
ground between the neutrality advocate and her critics. I’m not sure how the issue would be
reconfigured if reformulated in a noncognitivist framework.
7
On Objective List accounts of good, see Derek Parfit, Reasons and Persons (Oxford: Oxford
University Press, 1984), appendix.
8
. John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, rev. edition
1998); Rawls, Political Liberalism (New York, Columbia University Press, 2nd edition, 1996).
See also John Rawls, “Justice as Fairness, Political Not Metaphysical” (1985); Rawls, “The Idea
of an Overlapping Consensus” (1987) ; Rawls, “The Domain of the Political and Overlapping
Consensus” (1989); and Rawls, “The Idea of Public Reason Revisited” (1997). All of these
essays are reprinted in John Rawls, Collected Papersi\, ed. Samuel Freeman (Cambridge, MA:
Harvard University Press, 1999).
9
. Rawls, Political Liberalism, p. 136. The statement of the legitimacy ideal in the text differs in
a nontrivial way from Rawls’s formulation. In Rawls’s formulation, the legitimacy ideal only
applies to constitutional questions and basic matters of justice. Beyond these matters, it suffices
that political power is exercised in accordance with the procedures specified in a just constitution.
I set aside the complexities this formulation introduces that my formulation ignores. I don’t
believe they affect any arguments I go on to make.
10
In addition to the works of Rawls cited in note 8, see Burton Dreben, “On Rawls and Political
Liberalism,” and Charles Larmore, “Political Liberalism,” both in Samuel Freeman, ed., The
Cambridge Companion to Rawls (Cambridge: Cambridge University Press, 2003).
11
. It may seem that I am conflating the idea of the rational and the idea of the reasonable, two
notions Rawls distinguishes. For Rawls the rational person is the person of strategic prudence,
who takes the most efficient means to satisfy her goals. For Rawls, the reasonable person is
disposed to give due consideration to the morally valid claims of other persons. Rawls adds that
in the face of reasonable pluralism, the reasonable person is one who is not disposed to press her
comprehensive views as binding on all. “Reasonable” in Rawls’s usage does not then carry the
implication that a reasonable person registers and responds adequately to the reasons that are
there to be discerned. I use the term “rational” as equivalent to “fully reasonable” in the sense of
being perfectly responsive in one’s practical reasoning to the reasons there are. The fully
reasonable person makes no reasoning errors or any other cognitive errors in forming her moral
judgments; she is an ideal practical reasoner.